Corruption Perceptions Index
To measure the perceived levels of public sector corruption.
Combination of public, private sector and foundations
The index is used by a wide range of agencies as a measure of corruption. The results are widely publicised each year, making front page headlines in many countries. Some donors also use the index within their allocation models.
The CPI is a composite index, a poll of polls, drawing on corruption-related data from expert and business surveys carried out by a variety of independent institutions. The CPI is calculated using data from 17 different surveys or assessments produced by 13 organisations, this may change from year to year. All sources measure the overall extent of corruption (frequency and/or size of bribes) in the public and political sectors, and all sources provide a ranking of countries, i.e. include an assessment of multiple countries.This means methodological comparability across countries for any one input source.
As of 2011, Transparency International requires that three separate sources of information be available for a country in order for it to be included in the ranking.
Coutry Coverage: Globally but the number of countries varies from year to year: 177 Countries in 2013; 176 countries covered in 2012; and 183 countries covered in 2011.
Year Coverage: 1995 to present, updated annually.
In addition to Transparency International’s website, further information on the applied methodology can be obtained from:
Tel: + 49 30 3438200
Fax: + 49 30 3470 3912
Details of the questions asked by each of the data sources are available in the background paper on the website, released at the same time as the index. All data sources ask qualitative questions to determine the level of corruption. Note that some data sources (Economist Intelligence Unit, Freedom House) use exclusively external assessors based outside the country rated.
In previous editions of the CPI, the methodology drew on a country/territory’s rank position in the data sources to capture perceptions of corruption as compared with other countries/territories, as it was assumed that overall levels of corruption globally are unchanged from year to year. As of 2012 the CPI uses the raw scores given to any country/territory and then converts these raw scores to fit the CPI scale. As of 2012 the index is calculated using a simple average of standardized scores. More specifically, all thirteen sources are standardized by subtracting the mean of the data and dividing by the standard deviation (z-scores) and then rescaled to have a mean 45 and standard deviation 20. After the standardization, any values beyond the 0-100 scale are capped.
The format of the results were change in the 2012 to a 0 to 100 scale, with 0 indicating highly corrupt, and 100 indicating very clean. In the previous indices the scale runs from 0 to 10.0 with 10.0 being ‘highly clean’ and 0 being ‘highly corrupt’. A country's rank indicates its position relative to the other countries/territories included in the index.
The key purpose behind the index is to encourage countries governments, businesses, and civil society in countries towards the bottom of the index to do all they can to fight corruption and pull them up and away from the bottom of the Corruption Perceptions Index.
Furthermore, it can be used to illustrate the relationship between the extent of corruption and other important outcomes. For example the Corruption Perceptions Index can be plot against the UN’s Human Development Index, to discover the connection between corruption and the level of human development outcomes.
The previous versions of the index (1995-2011) does not allow for a country scores to be compared over time. This is because the index draws on a country's rank in the original data sources, rather than its score. A rank will always deliver relative positions. Those rankings are therefore a one off assessment. This procedure was changed as of 2012. A country's rank is now based on its raw score, thus in time the index may allow for the comparison of data over time.
Furthermore, the reliability differs between countries. Countries with a low number of sources and large differences in the evaluations provided by the sources (indicated by a wider confidence range) convey less reliability as to their score and ranking.
Note that this is a perception survey, thus the assumption is that corruption perception reflects real corruption experience.
CPI 2012 (Source: Transparency International - CPI 2012 l Interactive, 2012)